The mitochondria! antioxidant enzyme manganese-containing superoxide dismutase (MnSOD) has been shown to be a tumor suppressor in a wide variety of different cancer cell types. In this application, we propose to examine the mechanism of tumor suppression by MnSOD. We will examine which reactive oxygen or nitrogen species is involved. We propose that there are only three logical possibilities: superoxide radical, hydrogen peroxide, or nitric oxide radical. We will transfect the cDNA for proteins that affect these molecules into MnSOD over-expressing human breast cancer and glioma cell lines and determine the effect on the malignant phenotype. The other proteins that will be over-expressed include two forms of glutathione peroxidase, catalase, and nitric oxide synthase. Moreover, the effect of inhibiting the activity of these proteins will be determined. Inhibition of glutathione peroxidase, catalase, and nitric oxide synthase will be accomplished by chemical inhibitors as well as antisense oligodeoxynucleotides and siRNAs. We will also transfect glucose oxidase into cancer cells and examine the effect on the malignant phenotype. Glucose oxidase produces hydrogen peroxide when glucose is present as substrate and thus this experiment will examine further if hydrogen peroxide is the effector molecule. If we can determine the molecular effector of the growth inhibitory effect of MnSOD over-expression, then we can use that knowledge to develop new anticancer treatments based on MnSOD over-expression and inhibition of the effector removal pathway. For example, we already have data showing hydrogen peroxide is involved in the growth suppressive effect of MnSOD over-expression and so we plan to study the antitumor effect of MnSOD over-expression combined with inhibitors of hydrogen peroxide removal. In this way, we hope to be able to develop exciting new cancer therapies.